Teixeira, Ana Maria Miranda Botelho; (1996) Structural analysis of the biliary glycoprotein (BGP) binding site. Doctoral thesis (Ph.D), UCL (University College London).

Text out.pdf Download (22MB)

Abstract

The Biliary Glycoprotein (BGP) gene is a member of the Carcino-embryonic Antigen (CEA) Gene Family. It codes for at least 10 isoforms obtained by differential splicing of the BGP gene. Structurally they are members of the Immunoglobulin super family (IgSF) and function as homotypic adhesion molecules. It is clustered as CD66a and is mainly present in leukocytes and in theepitheliumof bile caniculi and colon. CHO transfectants expressing the BGPc isoform and soluble fusion protein chimeras (N-Fc, NA1B1-Fc and NA1B1A2-Fc) were used in binding assays where binding of these transfectants to the soluble domain proteins was tested. They were shown to bind in a homophilic manner suggesting that the N-terminal domain was involved in the adhesion process. BGP isoforms with an extracellular N-terminal domain only, followed by a transmembrane domain and a 72 or 9 amino acid cytoplasmic tail (BGPx, BGPx') have also been described. In order to ascertain the N-N domain nature of this adhesion a chimeric construct equivalent to the BGPx' isoform was constructed by a two step PCR method using primers containing the sequences of the end of the N-terminal domain followed by the beginning of the transmembrane domain, together with primers for either the beginning of the BGPc sequence or the end of the cytoplasmic tail. Sequencing of this construct matched the predicted sequence and a stable CHO-BGPx' cell line was established. Expression of this isoform at the cells surface was confirmed by immunofluorescence staining and western blot. Adhesion assays showed that this cell line was able to adhere to the N-Fc construct confirming the N-N terminal nature of this adhesion. To further localise the binding site a set of peptides spaning the C, C' and FG [beta] strands of the N-terminal domain were tested for their ability to inhibit adhesion. Both the C, CC' and FG peptides were able to partialy inhibit adhesion with the FG peptide being the most effective. This peptide was also able, at a higher concentration, to promote high levels of adhesion, using either transfected CHO-BGPc or untransfected CHO cells. These results indicate that this peptide was able to activate adhesion molecules present in the CHO cells probably integrins. To establish which amino acids might be involved in the BGP homophilic adhesion process, site directed mutagenesis on selected amino acids was performed using a PCR based method. Amino acids were selected based on a three dimensional model for the CEA molecule and the crystallographic data available for the rat CD2 molecule. After cloning, sequencing and transfection into Cos7 cells, the different mutant soluble protein constructs were purified and used in a binding assay to determine their effect on the BGP homophilic adhesion. Mutant proteins had either a single, double or triple amino acids substituted for alanine residues. Decreased adhesion was shown for single mutants V39 and V96, and for triple mutants S32Y34V39 and T87Q89I91. V39 and V96 are located in the CC and FG loops that connect the C to C and F to G [beta] strands respectively. Triple mutants S32Y34V39 and T87Q89I91 belong to the C and F [beta] strands respectively. Futher evidence for the importance of the F [beta] strand and FG loop came from the fact that one of the antibodies shown to block BGP homophilic adhesion was also shown to map to this region.

Download activity - last month

Export as CSVXMLJSON

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item